CN107852108B

CN107852108B - Eliminating commutation failure in hybrid HVDC systems

Info

Publication number: CN107852108B
Application number: CN201680039514.5A
Authority: CN
Inventors: 张小平; 雪映
Original assignee: University of Birmingham
Current assignee: University of Birmingham
Priority date: 2015-05-05
Filing date: 2016-04-11
Publication date: 2020-07-17
Anticipated expiration: 2036-04-11
Also published as: CN107852108A; US20180159421A1; US10305370B2; GB201507669D0; EP3292626B1; EP3292626A1; WO2016177991A1

Abstract

A line commutated converter L CC for a high voltage direct current, HVDC, power converter comprises at least one bridge circuit for connection to at least one terminal of a DC system, each bridge circuit comprising a plurality of arms and each arm being associated with a respective phase of an AC system, each arm comprising an upper and a lower thyristor connected in series, an associated branch extending between the upper and lower thyristors and at least one capacitor module for each phase.

Description

Eliminating commutation failure in hybrid HVDC systems

Technical Field

The present invention relates to power converters, and more particularly to line commutated converters (L CC) for use in High Voltage Direct Current (HVDC) systems and the control of faults therein.

Background

In High Voltage Direct Current (HVDC) power transmission systems Direct Current (DC) is used for large scale transmission of electric power. Since the power losses in a DC power transmission system are lower than in a corresponding AC power transmission system, DC is generally preferred for long distance power transmission. In addition, the cost of the long-distance DC transmission line is low. This is because DC transmission systems do not need to support three phases and have no skin effect, so DC requires less conductor area than AC.

In HVDC, high voltage AC must be converted to high voltage DC (rectified) before transmission, and the high voltage DC must then be reconverted to AC (inverted). typically a line commutated converter (L CC) or a Voltage Source Converter (VSC) is used for rectification and inversion. L CC systems are generally more popular than VSC systems because L CC can be used to transmit more power.

However, VSC systems typically use transistors that can be turned on and off, while L CC employs thyristors that can only be turned on (more precisely thyristor valves). The thyristors begin to conduct when forward biased, with their gate terminals receiving current triggers, and will continue to conduct until no longer forward biased.

In L CC systems, in addition, the current in the converter lags the voltage due to the time at which the commutation begins and the duration of the commutation, and the system consumes reactive power.

The present invention has been devised based on the above.

Disclosure of Invention

According to a first aspect of the invention there is provided a line commutated converter L cc. L CC for a high voltage direct current, HVDC, power converter comprising at least one bridge circuit for connection to at least one terminal of a DC system, each bridge circuit comprising a plurality of arms and each arm being associated with a respective phase of an AC system, each arm comprising an upper and a lower thyristor connected in series, an associated branch extending between the upper and lower thyristors and at least one capacitor module for each phase.

One advantage of the present invention is that commutation failures under single-phase and three-phase earth faults can be completely eliminated. The time required for commutation is also reduced.

In one embodiment, the capacitor module is operable to insert a first capacitor into the bridge circuit during a first commutation period. In the first commutation period, commutation is provided from a first initially conducting thyristor in the first initially conducting arm to a first subsequently conducting thyristor in the first subsequently conducting arm.

In one embodiment, the capacitor module is operable to insert a first capacitor into the first initially conducting arm to increase the effective commutation voltage. This reduces the current flowing through the first initially conducting thyristor and charges the capacitor.

In one embodiment, the capacitor module is operable to insert the first capacitor into the bridge circuit during the second commutation period. In the second commutation period, commutation is provided from a second initially conducting thyristor in a second initially conducting arm to a second subsequently conducting thyristor in a second subsequently conducting arm.

In one embodiment, the capacitor module is operable to insert the charged first capacitor into the second subsequently conducting arm to increase the effective commutation voltage. This increases the current through the second subsequently conducting thyristor and discharges the first capacitor.

In one embodiment, in the "push method", the first initially conducting thyristor and the first subsequently conducting thyristor are upper thyristors, and the second initially conducting thyristor and the second subsequently conducting thyristor are lower thyristors.

In one embodiment, in the "pull method", the first initially conducting thyristor and the first subsequently conducting thyristor are lower thyristors, and the second initially conducting thyristor and the second subsequently conducting thyristor are upper thyristors.

An advantage associated with the "push method" and the "pull method" is that the capacitor can be charged and discharged during the commutation period. This means that the capacitor voltages are approximately balanced.

In one embodiment, in a "push-pull method", during the first commutation period, the capacitor module is operable to insert a second capacitor into the first subsequently conducting arm to increase the effective commutation voltage. This increases the current through the first subsequently conducting thyristor and discharges the second capacitor.

The advantage of the "push-pull method" is that the insertion of two capacitors further increases the commutation voltage and means that the voltage of the inserted capacitor of each module can be about half the voltage of each inserted capacitor in the "push method" or the "pull method" and at the same time achieve similar commutation performance.

In one embodiment, during the second commutation period, the capacitor module is operable to insert a third capacitor into the second initially conducting arm to increase the effective commutation voltage. This reduces the current through the second initially conducting thyristor and charges the third capacitor.

In one embodiment, the capacitor module is operable to insert the first capacitor into the bridge circuit during the third commutation period. In a third commutation period commutation is provided from a third initially conducting thyristor in a third initially conducting arm to a third subsequently conducting thyristor in a third subsequently conducting arm.

In one embodiment, the capacitor module is operable to insert the first capacitor into the third initially conducting arm to increase the effective commutation voltage. This reduces the current flowing through the third initially conducting thyristor and charges the capacitor.

In one embodiment, the capacitor module is operable to insert the first capacitor into the bridge circuit during the fourth commutation period. In a fourth commutation period commutation is provided from a fourth initially conducting thyristor in a fourth initially conducting arm to a fourth subsequently conducting thyristor in a fourth subsequently conducting arm.

In one embodiment, the capacitor module is operable to insert the charged first capacitor into the fourth subsequently conducting arm to increase the effective commutation voltage. This increases the current through the fourth subsequently conducting thyristor and discharges the first capacitor.

In one embodiment, the first initially conducting thyristor, the first subsequently conducting thyristor, the fourth initially conducting thyristor, and the fourth subsequently conducting thyristor are upper thyristors. The second initially conducting thyristor, the second subsequently conducting thyristor, the third initially conducting thyristor and the third subsequently conducting thyristor are lower thyristors.

Another advantage of the push-pull method is that: the same capacitor is inserted in the primary conducting arm and the subsequent conducting arm for both the upper and lower thyristors. This means that the capacitor charges and discharges to a similar extent during the commutation period and the capacitor voltage is balanced.

Another advantage of the present invention is that independent fast reactive power control and tracking on the inverter side can be achieved by controlling the firing angle. With fast reactive power control and the proposed converter topology, a positive and negative reactive power exchange with the AC network on the inverter side can be achieved. This means that the turn-off angle may even be negative, which will output positive reactive power to the AC network. When the off angle is positive, the inverter absorbs reactive power from the AC network. When the off-angle decreases, the reactive power absorbed by the inverter from the AC network decreases. This results in cost savings because less reactive power support is required on the inverter side. Smaller converter transformer ratings, lower converter losses and fewer thyristor levels per valve can be achieved.

In one embodiment, each capacitor module is operable to insert a capacitor into a branch of the bridge circuit.

In one embodiment, each capacitor module is operable to insert a capacitor into an arm of the bridge circuit that is located above or below the associated branch.

The capacitor modules may be connected in series to form a plurality of modular capacitors (MMCs).

In one embodiment, at least one of the capacitor modules is a full bridge circuit.

In one embodiment, the or each full bridge circuit is configured to insert a capacitor into the arm in either polarity.

The advantage of using a full bridge circuit is that the circuit is simple and requires a small number of semiconductor components.

In one embodiment, at least one of the capacitor modules is a dual clamping circuit.

In one embodiment, the or each dual clamping circuit is configured to insert two capacitors into the limb in either polarity.

One advantage of using a dual clamp circuit is that: the dual clamp circuit may have a higher output voltage or lower losses with the same voltage output as the full bridge circuit.

In one embodiment, at least one of the capacitor modules is a five-stage cross-connect circuit.

In one embodiment, the or each five-stage cross-connect circuit is configured to insert one or two capacitors into the arm in either polarity.

The advantage of using a five-stage cross-connect circuit is: as with the dual clamp circuit, this circuit can have a higher output voltage or lower losses if the same voltage is output as the full bridge circuit. In addition to this, a five-stage cross-connect circuit has more switching states, which means that zero, one or two of the capacitors can be inserted into the circuit in either direction.

In one embodiment, at least one of the capacitor modules is a hybrid commutation circuit.

In one embodiment, the or each hybrid commutation circuit is configured to selectively insert the first capacitor in a first polarity, insert both the first capacitor and the second capacitor in the first polarity or insert the first capacitor in the second polarity.

One advantage of the hybrid commutation circuit is that it has some of the functions of two full-bridge circuits connected together, but with a smaller number of switching devices.

According to a second aspect of the invention there is provided a method of operating a line commutated converter L CC for a high voltage direct current, HVDC, converter, said method operating L CC, the L CC comprising at least one bridge circuit for connection to at least one terminal of a DC system, each bridge circuit comprising a plurality of arms and each arm being associated with a respective phase of an AC system.

In one embodiment, the capacitor module inserts a first capacitor into the bridge circuit during a first commutation period. In the first commutation period, commutation is provided from a first initially conducting thyristor in the first initially conducting arm to a first subsequently conducting thyristor in the first subsequently conducting arm.

In one embodiment, the capacitor module inserts a first capacitor into the first initially conducting arm to increase the effective commutation voltage. This reduces the current flowing through the first initially conducting thyristor and charges the first capacitor.

In one embodiment, the capacitor module inserts a first capacitor into the bridge circuit during the second commutation period. During a second commutation period, commutation is provided from a second initially conducting thyristor in a second initially conducting arm to a second subsequently conducting thyristor in a second subsequently conducting arm.

In one embodiment, the capacitor module inserts a charged first capacitor into the second subsequently conducting arm to increase the effective commutation voltage. This increases the current through the second subsequently conducting thyristor and discharges the first capacitor.

In one embodiment, the method further comprises controlling the firing angle, which is the phase angle between the point at which the effective commutation voltage becomes positive and the point at which the thyristor is fired. The firing angle is controlled by controlling the timing of the firing of the thyristor. This makes it possible to control the exchange of reactive power with the AC system. One advantage of this is that positive and negative reactive power exchange with the AC network on the inverter side can be achieved. When the off-angle decreases, the reactive power absorbed by the inverter from the AC network decreases. This results in cost savings because less reactive power support is required on the inverter side. Smaller converter transformer ratings, lower converter losses and a lower number of thyristor levels per valve can be achieved.

In one embodiment, the method further comprises controlling the timing of firing of thyristors of the converter operating as an inverter to provide a varying turn-off angle for the inverter to provide controllable reactive power to the AC system. The timing can be controlled so that the off angle is negative. The off-angle is the phase angle between the end of the commutation period and the point at which the natural commutation voltage becomes negative. The natural commutation voltage is the commutation voltage without modification (without the use of a capacitor).

In one embodiment, the method further comprises controlling the timing of firing of thyristors of said converter operating as a rectifier to provide a varying firing angle for said rectifier to provide controllable reactive power to the AC system. The timing may be controlled such that the firing angle is negative, thereby providing positive reactive power to the AC system.

Physically, the state of an inverter with a negative turn-off angle is similar to the state of a rectifier with a negative firing angle.

In an embodiment, the effective commutation voltage prevents commutation failure of the HVDC during a fault of the AC system.

In an embodiment wherein the L CC converter is used as a rectifier for converting from AC to DC and as an inverter for converting from DC to AC, wherein during an AC system fault the active commutation voltage is controlled to provide controllable reactive power support to the AC system, whereby the rectifier and the inverter each provide positive reactive power to the AC system, wherein there is negative firing angle control for the rectifier and negative turn-off angle control for the inverter.

Drawings

Fig. 1 is a circuit diagram showing a three-phase L CC inverter according to an embodiment of the present invention.

Fig. 2a is a circuit diagram of an embodiment of the invention showing a "push method" using capacitor insertion.

Fig. 2b is a circuit diagram showing an embodiment of the invention using a "push method" of capacitor insertion at a different point in the commutation period than fig. 2 a.

Fig. 3a is a circuit diagram illustrating an embodiment of the invention using the "pull method" of capacitor insertion.

Fig. 3b is a circuit diagram showing an embodiment of the invention using a "pull method" of capacitor insertion at a different point in the commutation period than fig. 3 a.

Fig. 4 is a circuit diagram illustrating an embodiment of the present invention using the "push-pull method" of capacitor insertion.

Fig. 5 shows a time-dependent curve of the thyristor current and the capacitor voltage.

Fig. 6a and 6b are schematic diagrams showing a capacitor module.

Fig. 6c is a circuit diagram of a full bridge circuit that may be used for the SMC module.

Fig. 7a is a circuit diagram of a dual clamping circuit as an alternative circuit that may be used in an SMC module.

Fig. 7b is a circuit diagram of a five-stage cross-connect circuit as an alternative circuit that may be used for the SMC module.

Fig. 7c is a circuit diagram of a hybrid commutation circuit as an alternative circuit that can be used for the SMC module.

Fig. 8 is a circuit diagram of a portion of an alternative three-phase L CC inverter.

Detailed Description

Referring to fig. 1, a three-phase L CC inverter 2 is shown, L CC inverter 2 converts power from DC to AC power from DC transmission lines (not shown) connected to first and

second DC terminals

4, 6 and is delivered to a three-phase AC system 3, according to an embodiment of the invention, first DC terminal 4 directs current into inverter 2, while second DC terminal 6 directs current out of inverter 2-the DC current remains substantially constant over time.

The inverter 2 has a 12-pulse arrangement, in which an upper 6-pulse bridge 8 and a lower 6-pulse bridge 10 are connected in series. The first DC terminal 4 is connected to the upper 6-pulse bridge 8 and the lower 6-pulse bridge 10 is connected to the second DC terminal 6. Although the invention is described herein with respect to a 12-pulse bridge, it should be understood that the invention may be implemented with other bridge circuits. For example, the invention may be used with any 6 k-pulse bridge arrangement, where k is 1, 2, 3, 4.

The upper 6-pulse bridge 8 comprises three

parallel arms

12, 14, 16 (one for each of the three phases of the AC system 3), each comprising an

upper thyristor

12a, 14a, 16a and a

lower thyristor

12b, 14b, 16b connected in series. The lower 6-pulse bridge 10 also comprises three

parallel arms

18, 20, 22 (one for each of the three phases of the AC system 3), each comprising an

upper thyristor

18a, 20a and a

lower thyristor

18b, 20b connected in series. All thyristors are connected in their polarity so that they can conduct current from the DC system when switched on.

In each 6-

pulse bridge

8, 10, each parallel arm 12, M, 16, 18, 20, 22 comprises a branch connection 23a-23c, 25a-25c to a respective one of the three phases of the AC system 3. Each branch connection 23a-23c, 25a-25c is connected to the parallel arms at a point between the upper 12a, 14a, 16a and lower 12b, 14b, 16b thyristors of the

parallel arms

12, 14, 16, 18, 20, 22.

For the upper 6-pulse bridge 8 the branch connections 23a-23c are connected to the AC system 3 via wye-wye (star-star) transformers 25, whereas for the lower 6-pulse bridge 10 the branch connections 25a-25c are connected to the AC system 3 via delta-wye (delta-star) transformers 27.

In the present invention, each branch connection 23a-23c, 25a-25c comprises a capacitor module 24a-24c, 26a-26c into which a capacitor can be inserted into the branch connection 23a-23c, 25a-25 c. This is described in more detail below with reference to fig. 3 and 4.

The ideal operation of the inverter 2 is briefly described in the following paragraphs for the upper 6-pulse bridge 8. It will be appreciated that the lower 6-pulse bridge 10 operates in the same manner as the upper 6-pulse bridge 8. The effect of the capacitor modules 24a-24c, 26a-26c is not considered at this time.

In an ideal inverter, when no commutation occurs, a first one of the

upper thyristors

12a, 14a, 16a conducts DC current. At the same time, the first of the

lower thyristors

12b, 14b, 16b of the

parallel arms

12, 14, 16, which is different from the conducting upper thyristor, also conducts a DC current. This means that the current flowing through two of the three phases is equal to the DC current.

Shortly thereafter, a commutation period begins, wherein commutation is provided from an initially conducting thyristor in an initially conducting arm to a subsequently conducting thyristor in a subsequently conducting arm. This means that at the beginning of the commutation period, the next upper or lower thyristor is triggered (receives current triggering at its gate terminal) and starts to conduct. At this time, a voltage difference, a so-called effective commutation voltage, exists between the two commutation phases. In order for commutation to occur, the effective commutation voltage must be positive so that the voltage for the lower thyristor subsequently conducting phase is greater than the voltage for the initial conducting phase and the voltage for the upper thyristor subsequently conducting phase is less than the voltage for the initial conducting phase. This means that the current in the next upper or lower thyristor starts to increase and the current in the corresponding (upper or lower) first thyristor starts to decrease. This continues until the current in the first thyristor is below the holding current of the thyristor, and the commutation period ends. For systems without capacitors, the commutation is driven only by the effective commutation voltage.

The commutation period represents a short overlap period during which current passes through both the first and second thyristors. This is described in more detail below with reference to fig. 2a to 2 c. The commutation period ends when the first thyristor turns off and current passes through only one of the

upper thyristors

12a, 14a, 16a and one of the

lower thyristors

12b, 14b, 16 b. The overlap angle mu is equal to the phase angle of the current flowing through both the first and second thyristors.

Conventional systems can be controlled by controlling the firing angle α, the firing angle α being the phase angle between the point in the commutation period where the effective commutation voltage becomes positive and the point at which the thyristor is fired.

It is commonly referred to as the off-angle γ, which is the phase angle between the end of the commutation period and the point in the commutation period where the effective commutation voltage becomes negative. The turn-off angle may also be defined by the relation given below.

γ＝180°-μ-α

Where γ is the off angle, μ is the overlap angle, and α is the firing angle.

The above switching process continues and the thyristors are switched in a repeated sequence so that three phase AC is generated in the AC system 3.

In the following discussion, capacitors are used to modify (e.g., increase) the effective commutation voltage during various portions of the commutation period. It can be assumed that the effective commutation voltage is equal to the natural commutation voltage plus the voltage of the inserted capacitor. Hence, where it is necessary to represent an effective commutation voltage (such as in relation to defining a firing angle or a turn-off angle) without such a modification (without the use of a capacitor), it is referred to herein as a natural commutation voltage.

The operation of the inverter 2 may be interrupted due to an AC fault. In an AC fault, one or more of the AC phases may be shorted to ground. When the thyristor corresponding to the short-circuited AC phase is triggered, a large current flows through the thyristor. The value of this current is still greater than the holding current of the thyristor even after the next thyristor is triggered. This means that the thyristors do not turn off, resulting in system commutation failure. The system often needs to be restarted due to commutation failure. The present invention uses the capacitor modules 24a-24c, 26a-26c described above with reference to fig. 1 to eliminate such commutation failures.

Fig. 2a to 4 show a part of the inverter circuit 2 (see fig. 1) during a commutation period in which two thyristors are conducting simultaneously. Fig. 2a to 4 likewise show only the upper 6-pulse bridge 8. However, it will be appreciated that the operation of the lower 6-pulse bridge 10 is the same as that of the upper 6-pulse bridge 8. Examples of capacitor modules 24a-24c, 26a-26c for inserting capacitors into the commutation circuit loop are given below with reference to fig. 6a to 6c and fig. 7a to 7 c.

The start of the commutation period from thyristor TY 216 a to thyristor TY412a is represented by a point in the cycle shown in fig. 2a, 3a and 4. At this point TY412a has just been triggered. This means that TY 216 a will continue to conduct until its current drops below the holding current of the thyristor. Fig. 2b and 3b show different points in the commutation periods of fig. 2a, 3a and 4, as described below.

In all cases described below with respect to fig. 2a to 4, the insertion of the capacitor acts together with the voltage difference between the two commutation phases to effect commutation by increasing the effective commutation voltage. In addition to minimizing the effects of AC faults, the insertion of the capacitor reduces the time it takes to complete the commutation period. Although neither of the figures shows an AC fault, it will be appreciated that the capacitor insertion approach eliminates commutation failures during a fault event.

Referring to fig. 2a, one embodiment of the present invention using a "push method" of capacitor insertion is shown. In the present embodiment, at the instant TY412a turns on through its gate terminal, capacitor CapYc34c is inserted into c-phase branch 23c, so that its positive plate is connected to TY 216 a. CapYc34c reduces the current through TY 216 a and "pushes" the current to TY 4. CapYc34c charges during this commutation period. This continues until the current through TY 216 a reaches a value less than the thyristor holding current, and TY 216 a turns off.

Referring to fig. 2b, an embodiment of the invention is shown using a "push method" of capacitor insertion at a different point in the commutation period than fig. 2 a. In fig. 2b TY314b is on and TY 516 b has just been triggered. At this time, the charged CapYc34c is inserted into c-phase branch 23c again in the same direction as in fig. 2 a. CapYc34c increases the current through TY 516 b and thus decreases the current through TY314 b. This continues until the current through TY314b reaches a value less than the thyristor holding current, and TY314b turns off. CapYc34c discharges during this commutation period.

The "push method" is performed at two other points in time of the commutation period (not shown) by inserting a capacitor CapYa into the a-phase branch 23a to decrease the current through the thyristor TY412a or to increase the current through the thyristor TY 112 b. The method is also performed at two other points in time of the commutation period (not shown) by inserting CapYb into the b-phase branch 23b to reduce the current through thyristor TY 614 a and increase the current through thyristor TY314 b.

Each capacitor has current in both directions during the entire commutation period. This means that when the "push method" is performed, the capacitor charges and discharges to a similar extent and the capacitor voltages are approximately balanced.

Referring to fig. 3a, one embodiment of the present invention using a "pull method" of capacitor insertion is shown. In this embodiment, at the instant TY412a is triggered, the charged capacitor CapYa34a is inserted into phase a branch 23a, so that its negative plate is connected to TY412 a. CapYa34a "pulls" the current to TY412a, increasing the current through TY412a, and thus decreasing the current through TY 2. This continues until the current through TY 216 a reaches a value less than the thyristor holding current, and TY 216 a turns off. During this commutation process, the CapYa34a discharges, thereby providing current to the AC system 3.

Referring to fig. 3b, an embodiment of the invention using the "pull method" of capacitor insertion at a different point in the commutation period than fig. 3a is shown. In fig. 3b TY 112b is on and TY314b is triggered. At this time, the CapYa34a is inserted into the phase a branch 23a in the same direction as in fig. 2 b. CapYa34a resulted in a decrease in current through TY 112 b. This continues until the current through TY 112b reaches a value less than the thyristor holding current, and TY 112b turns off. The CapYa34a charges during this commutation period.

The "pull method" is performed at two other points in time of the commutation period (not shown) by inserting a capacitor CapYb into the b-phase branch 23b to increase the current through the thyristor TY 614 a or to decrease the current through the thyristor TY314 b. The method is performed at the other two points in time of the commutation period by inserting CapYb 34c into the c-phase branch 23c to increase the current through thyristor TY 216 a or to increase the current through thyristor TY 516 b.

Each capacitor has current in both directions during the entire commutation period. This means that when the "pull method" is performed, the capacitor charges and discharges to a similar extent and the capacitor voltage is approximately balanced.

Referring to fig. 4, one embodiment of the present invention using the "push-pull method" is shown. In this example, at the instant TY412a is triggered, CapYc34c is inserted into the c-phase branch 23c, so that its positive plate is connected to TY 216 a. At the same time, CapYa34 b was inserted into phase a branch 23a, so that its negative plate was connected to TY412 a. This insertion increases the effective commutation voltage between phase c and phase a by the sum of the two capacitor voltages. This helps to significantly reduce the current through TY 216 a until the current reaches a value less than the thyristor holding current and TY 216 a turns off. During this process, CapYc34c charges and CapYa34a discharges.

In the "push-pull method", at other points in the cycle, CapYa34a, CapYb, and CapYc34c are inserted into the branches, as described above for the "push method" and the "pull method".

One advantage of the "push-pull method" is that: during all commutation periods, each capacitor has a current in both directions when inserted near one of the

upper thyristors

12a, 14a, 16 a. In addition to this, each capacitor has a current in both directions when inserted near one of the

lower thyristors

12b, 14b, 16 b. Since the DC current is approximately constant, the capacitor is discharged to the same extent throughout the cycle. That is, for the "push-pull method", the system is balanced.

While the above-described "push method" and "pull method" provide useful illustrations of system operation, it will be appreciated that there is a great degree of similarity between the two methods. In both cases, the capacitor module inserts a capacitor into the circuit during the commutation period. Otherwise, the two methods are similar in the following description.

In the "push method" with respect to the upper thyristor (as shown in fig. 2 a) and the "pull method" with respect to the lower thyristor (as shown in fig. 3 b), the capacitor module inserts a capacitor into the initially conducting arm to increase the effective commutation voltage. This reduces the current through the initially conducting thyristor and charges the capacitor.

In the "push method" with respect to the lower thyristor (as shown in fig. 2 b) and the "pull method" with respect to the upper thyristor (as shown in fig. 3 a), the capacitor module inserts a capacitor into the subsequently conducting arm to increase the effective commutation voltage. This increases the current through the subsequently conducting thyristor and discharges the capacitor.

Referring to fig. 5, a plot 36 of thyristor current versus time and a plot 46 of CapYc voltage versus time are shown for TY 238, TY 340, TY 442 and TY 544. These graphs represent the "push method" or "push-pull method" as described above with reference to fig. 2a and 4.

At a first time 48, a commutation period from TY2 to TY4 begins. As described above with reference to fig. 2a, 3a and 4. At a second time 50, the commutation period from TY2 to TY4 ends. Between first time 48 and second time 50, CapYc is inserted into the c-phase branch such that its positive terminal is connected to TY2 (to "push" the current to TY 4). CapYc is charged and its voltage rises from V _ low to V _ high. The current in TY2 decreases from DC value to zero and the current in TY4 increases from zero to DC value.

At a third time 52 the commutation period from TY3 to TY5 starts, and at a fourth time 54 the commutation period from TY3 to TY5 ends. Between third time 52 and fourth time 54, CapYc is inserted into the c-phase branch in the same direction as described in the previous paragraph. However, at this time the current in the c-phase branch is in the opposite direction and CapYc discharges. This causes the voltage across CapYc to return from V _ high to V _ low. The current in TY3 decreases from DC value to zero and the current in TY5 increases from zero to DC value.

In a conventional L CC system (without the capacitor insertion described above), the turn-off angle (defined above) related to the thyristor turn-off time cannot be too small to ensure that commutation margin remains for possible disturbances on the AC side.

By the method described herein commutation failures can be eliminated and the turn-off angle can take a wider range of values. This means that the firing angle can be controlled within a larger range of values, allowing an efficient control of the exchange of reactive power with the AC system. This allows for a significant reduction in the size of the reactive power support equipment, smaller converter transformer ratings, lower converter losses, and fewer thyristor stages. Furthermore, the firing angle may be controlled such that the off-angle of the converter is negative when operating as an inverter (when DC power is converted to AC), thereby providing reactive power to the AC system. Again, the firing angle is controlled by controlling the timing of the firing of the thyristors. This ability to control the firing angle and/or the turn-off angle means that if the AC system fails, the effective commutation voltage can be controlled to provide reactive power to the AC system. This results in a reduction of the required reactive power support, allowing a reduction of the size of the AC filter, a smaller converter transformer rating and lower converter losses.

In addition to the above, a smaller turn-off angle means that the AC voltage can be lower for a given DC voltage level, and the thyristor can be rated for a lower voltage.

Referring to fig. 6a and 6b, schematic diagrams are shown showing that the capacitor module 24a may be a Single Modular Capacitor (SMC)28 or a series of

connected SMCs

28a, 28b, 28n forming a multi-module capacitor (MMC). The same is true for the

other capacitor modules

24b, 24c, 26a, 26b, 26c (see fig. 1).

Referring to fig. 6c, a circuit diagram of the SMC module 28 is shown that allows the capacitor 34 to be inserted into the branch connection 23a in either polarity. In the present embodiment, the SMC 28 constituting the capacitor module 24a is a full bridge circuit 29. The full bridge circuit 29 comprises four Insulated Gate Bipolar Transistors (IGBTs) 30a, 30b, 30c, 30d, each equipped with an integrated

anti-parallel diode

32a, 32b, 32c, 32 d. It will be appreciated that similar SMC modules may be used for the

capacitor modules

24b, 24c, 26a-26c in the

branch connections

23b, 23c, 25a-25 c.

Turning on the

IGBTs

30a, 30c and turning off the

IGBTs

30a, 30d causes the capacitor 34a to switch into the circuit in one direction, while turning on the

IGBTs

30a, 30d and turning off the

IGBTs

30b, 30c causes the capacitor 34a to switch into the circuit in the opposite direction. Turning on the

IGBTs

30a, 30c and turning off the

IGBTs

30b, 30d, or turning on the IGBTs 30b, 30d and turning off the

IGBTs

30a, 30c causes the capacitor 34a to be bypassed.

The advantage of using the full bridge circuit 29 is that the circuit is simple and requires a small number of semiconductor components.

Referring to fig. 7a, a circuit diagram of a dual clamp circuit 129 is shown as an alternative circuit that may be used for the SMC module 28. The dual clamp circuit 129 allows two

capacitors

134a, 134b to be inserted into one of the branch connections 23a-23c, 25a-25c in either polarity. The dual clamping circuit 129 comprises five IGBTs 130a-130e, each equipped with an integrated anti-parallel diode 132a-132e and two

further diodes

132e, 132 f.

Turning on the

IGBTs

130a, 130c, 130e and turning off the

IGBTs

130a, 130d causes the

capacitors

134a, 134b to switch into the circuit in one direction, while turning on the

IGBTs

130a, 130d and turning off the

IGBTs

130b, 130c, 130e causes the

capacitors

134a, 134b to switch into the circuit in the opposite direction. Turning on the

IGBTs

130a, 130d, 130e and turning off the

IGBTs

130b, 130c causes the

capacitors

134a, 134b to be bypassed.

One advantage of using the dual clamp circuit 129 is that: the dual clamp circuit may have a higher output voltage or lower losses with the same voltage output as the full bridge circuit. In addition, dual clamp circuit 129 has the ability to block fault currents.

Referring to fig. 7b, a circuit diagram of a five-stage cross-connect circuit 229 is shown, which is an alternative circuit that may be used for the SMC module 28. The five-stage cross-connect circuit 229 allows two

capacitors

234a, 234b to be inserted into one of the branch connections 23a-23c, 25a-25c in either polarity. The five-stage cross-connect circuit includes six IGBTs 230a-230f, each equipped with an integrated anti-parallel diode 232a-232 f.

Turning on the

IGBTs

230a, 230c, 230e and turning off the

IGBTs

230a, 230d, 230f causes the

capacitors

234a, 234b to switch into the circuit in one direction, while turning on the

IGBTs

230a, 230d, 230f and turning off the

IGBTs

230b, 230c, 230e causes the

capacitors

234a, 234b to switch into the circuit in the opposite direction. Turning on the

IGBTs

230a, 230d, 230e and turning off the

IGBTs

230b, 230c, 230f, or turning on the IGBTs 230b, 230c, 230f and turning off the

IGBTs

230a, 230d, 230e causes the

capacitors

234a, 234b to be bypassed.

The advantage of using the five-stage cross-connect circuit 229 is: as with the dual clamp circuit 129, this circuit can have a higher output voltage or lower losses if the same voltage is output as the full bridge circuit. In addition to this, the five-stage cross-connect circuit 229 has more switch states, which means that zero, one, or two of the

capacitors

234a, 234b can be inserted into the circuit in either direction.

Referring to fig. 7c, a circuit diagram of a hybrid commutation circuit 329 is shown, the hybrid commutation circuit 329 being another alternative circuit that may be used in the SMC module 28. The hybrid commutation circuit 329 allows a capacitor 334a (in either polarity) to be inserted and another capacitor 334b (in a single polarity) to be inserted into one of the branch connections 23a-23c, 25a-25 c. The hybrid commutation circuit 329 includes six IGBTs 330a-330f, each equipped with an integrated anti-parallel diode 332a-332 f.

Turning on the IGBTs 330b, 330c, 330e and turning off the

IGBTs

330a, 330d, 330f causes the

capacitors

334a, 334b to switch into the circuit in one direction, while turning on the

IGBTs

330a, 330d, 330f and turning off the

IGBTs

330b, 330c, 330e causes the capacitor 334a to switch into the circuit in the opposite direction. Turning on the

IGBTs

330a, 330c, 330f and turning off the

IGBTs

330b, 330d, 330e, or turning on the IGBTs 330b, 330d, 330f and turning off the

IGBTs

330a, 330c, 330e causes the capacitors to be bypassed.

One advantage of the hybrid commutation circuit 329 is that it has some of the functionality of two full-bridge circuits connected together, but has a smaller number of switching devices.

It will be appreciated that there may also be other alternative capacitor modules that may be used to insert capacitors into the circuit.

Referring to fig. 8, an alternative arrangement for a portion of a three-phase L CC inverter 102 is shown, the arrangement being the same as that described above with respect to fig. 1, but having two

capacitor modules

112a, 112b, 114a, 114b, 116a, 116b connected in each of the three

parallel arms

112, 114, 116.

L CC inverter 102 operates in the same manner as described above with respect to FIG. 1 and implements the same circuits during commutation as those shown above with respect to FIGS. 2 a-4 to ensure that the charging and discharging of the capacitors is balanced for L CC inverter 102, this arrangement is preferably used to implement the push-pull method as described above.

It will be appreciated that the invention also has other alternative embodiments in which the capacitor modules are connected at different locations. For example, in fig. 8, the positions of the capacitor module and the thyristor may be switched.

The above description explains the proposed control method using the converter shown in fig. 1 operating as an inverter as an example. However, the description and control principles apply to converters operating as rectifiers. In the case of a rectifier, the timing of the thyristor firing may be controlled such that a varying firing angle of the rectifier may be achieved to provide controllable reactive power as described (the firing angle for the rectifier may even be negative) to the AC system to provide positive reactive power to the AC system (i.e. the thyristor is fired before the natural commutation voltage becomes positive).

Claims

1. A line commutated converter L CC for a high voltage direct current, HVDC, power converter, the L CC comprising at least one bridge circuit for connection to at least one terminal of a DC system, wherein each bridge circuit comprises a plurality of arms, each arm being connected to a respective phase of an AC system, each arm comprising:

upper and lower thyristors connected in series;

a branch connection connected to a point between the upper and lower thyristors; and

at least one capacitor module for each phase, the capacitor modules being included in the arm or the branch connection and operable to insert a capacitor in either polarity into a respective arm of a bridge circuit.

2. The L CC of claim 1, wherein the capacitor module is operable to insert a first capacitor into the bridge circuit during a first commutation period, wherein commutation is provided from a first initially conducting thyristor in a first initially conducting arm to a first subsequently conducting thyristor in a first subsequently conducting arm.

3. The L CC of claim 2, wherein the capacitor module is operable to insert a first capacitor into a first initially conducting arm to increase an effective commutation voltage to reduce current flowing through a first initially conducting thyristor and charge the first capacitor.

4. L CC according to claim 2 or 3, wherein the capacitor module is operable to insert a first capacitor into the bridge circuit during a second commutation period, wherein commutation is provided from a second initially conducting thyristor in a second initially conducting arm to a second subsequently conducting thyristor in a second subsequently conducting arm.

5. The L CC of claim 4, wherein the capacitor module is operable to insert a charged first capacitor into a second subsequently conducting arm to increase an effective commutation voltage to increase current through a second subsequently conducting thyristor and discharge the first capacitor.

6. The L CC of claim 5, wherein, in the push method, the first initially conducting thyristor and the first subsequently conducting thyristor are upper thyristors and the second initially conducting thyristor and the second subsequently conducting thyristor are lower thyristors.

7. L CC according to claim 5, wherein in the "pull method" the first initially conducting thyristor and the first subsequently conducting thyristor are lower thyristors and the second initially conducting thyristor and the second subsequently conducting thyristor are upper thyristors.

8. The L CC of claim 5, wherein, in a "push-pull method", during a first commutation period, the capacitor module is operable to insert a second capacitor into a first subsequently conducting arm to increase the effective commutation voltage, thereby increasing the current flowing through the first subsequently conducting thyristor and discharging the second capacitor.

9. The L CC of claim 8, wherein during a second commutation period, the capacitor module is operable to insert a third capacitor into the second initially conducting arm to increase the effective commutation voltage, thereby reducing the current flowing through the second initially conducting thyristor and charging the third capacitor.

10. The L CC of claim 8, wherein the capacitor module is operable to insert the first capacitor into the bridge circuit during a third commutation period, wherein commutation is provided from a third initially conducting thyristor in a third initially conducting arm to a third subsequently conducting thyristor in a third subsequently conducting arm.

11. The L CC of claim 10, wherein the capacitor module is operable to insert a first capacitor into a third initially conducting arm to increase an effective commutation voltage to reduce current flowing through a third initially conducting thyristor and charge the first capacitor.

12. The L CC of claim 11, wherein the capacitor module is operable to insert the first capacitor into the bridge circuit during a fourth commutation period, wherein commutation is provided from a fourth initially conducting thyristor in a fourth initially conducting arm to a fourth subsequently conducting thyristor in a fourth subsequently conducting arm.

13. The L CC of claim 12, wherein the capacitor module is operable to insert the charged first capacitor into a fourth subsequently conducting arm to increase an effective commutation voltage to increase a current through a fourth subsequently conducting thyristor and discharge the first capacitor.

14. The L CC of claim 13, wherein:

the first initial conducting thyristor, the first subsequent conducting thyristor, the fourth initial conducting thyristor and the fourth subsequent conducting thyristor are upper thyristors; and

the second initially conducting thyristor, the second subsequently conducting thyristor, the third initially conducting thyristor and the third subsequently conducting thyristor are lower thyristors.

15. The L CC of claim 1, wherein at least one of the capacitor modules is one of a full bridge circuit, a double clamp circuit, a five-stage cross-connect circuit, and a hybrid commutation circuit.

16. A method of operating a line commutated converter L CC for a high voltage direct current, HVDC, converter, wherein the L CC comprises at least one bridge circuit for connection to at least one terminal of a DC system, wherein each bridge circuit comprises a plurality of arms, each arm being connected to a respective phase of an AC system, each arm comprising:

upper and lower thyristors connected in series;

at least one capacitor module for each phase, the capacitor modules included in the arm or the branch connection, the method comprising:

capacitors are inserted by the capacitor modules into the respective arms of the bridge circuit in either polarity.

17. The method of claim 16, wherein the capacitor module inserts a first capacitor into the bridge circuit during a first commutation period, wherein commutation is provided from a first initially conducting thyristor in a first initially conducting arm to a first subsequently conducting thyristor in a first subsequently conducting arm.

18. The method of claim 17, wherein the capacitor module inserts a first capacitor into the first initially conducting arm to increase the effective commutation voltage to reduce current flowing through the first initially conducting thyristor and charge the first capacitor.

19. The method of claim 18, wherein the capacitor module inserts the first capacitor into the bridge circuit during a second commutation period, wherein commutation is provided from a second initially conducting thyristor in a second initially conducting arm to a second subsequently conducting thyristor in a second subsequently conducting arm.

20. The method of claim 19, wherein the capacitor module inserts the charged first capacitor into the second subsequently conducting arm to increase the effective commutation voltage to increase the current flowing through the second subsequently conducting thyristor and discharge the first capacitor.

21. The method of any of claims 18 to 20, further comprising controlling a firing angle, wherein:

the firing angle is the phase angle between the point at which the natural commutation voltage becomes positive and the point at which the thyristor is fired; and

the firing angle is controlled by controlling the timing of the firing of the thyristor,

thereby controlling the exchange of reactive power with the AC system.

22. The method of claim 21, further comprising: controlling the timing of firing of thyristors of the converter operating as an inverter to provide a varying turn-off angle for the inverter to provide controllable reactive power to an AC system, wherein:

the off-angle is the phase angle between the end of the commutation period and the point at which the natural commutation voltage becomes negative.

23. The method of claim 22, further comprising controlling timing of firing of thyristors such that the turn-off angle is negative.

24. The method of claim 18, wherein during an AC system fault, the effective commutation voltage prevents commutation failure of the HVDC.